The potential of dynamic indicator in development of the bankruptcy prediction models: the case of construction companies

Abstract

The current development of bankruptcy models usually goes in the direction of testing different classification algorithms, while the potential hidden in financial indicators is given less attention. Their analysis is often only restricted to the comparison between their respective statuses in bankrupt and healthy companies, while the dynamics of the indicators, i.e. the change in their values in time, is not paid much attention. The aim or our research is to analyse partial potential of financial ratios for predicting bankruptcy. Twentyeight indicators were examined in a sample of 1,355 construction companies operating in the Czech Republic, as well as their development over the past five periods. A nonparametric chisquare test was used to evaluate the significance of predictors. The variables were categorised for the application of the test. Our research confirmed the assumption as to the importance of using the indicators in dynamic (change) form. Indicators that are significant only in their change form were identified. Moreover, the use of the dynamic form of the indicators can increase the significance of the bankruptcy model. This was tested by using the stepwise version of linear discrimination analysis.The current development of bankruptcy models usually goes in the direction of testing different classification algorithms, while the potential hidden in financial indicators is given less attention. Their analysis is often only restricted to the comparison between their respective statuses in bankrupt and healthy companies, while the dynamics of the indicators, i.e. the change in their values in time, is not paid much attention. The aim or our research is to analyse partial potential of financial ratios for predicting bankruptcy. Twentyeight indicators were examined in a sample of 1,355 construction companies operating in the Czech Republic, as well as their development over the past five periods. A nonparametric chisquare test was used to evaluate the significance of predictors. The variables were categorised for the application of the test. Our research confirmed the assumption as to the importance of using the indicators in dynamic (change) form. Indicators that are significant only in their change form were identified. Moreover, the use of the dynamic form of the indicators can increase the significance of the bankruptcy model. This was tested by using the stepwise version of linear discrimination analysis.